The objective of this proposal is to develop the synthetic potential of the photocyclization reaction in which the ultraviolet irradiation of o-methoxystilbenes in the absence of dissolved oxygen gives 4a-methoxy-4a, 4b-dihydrophenanthrenes as transient intermediates that subsequently can undergo elimination of methanol to give phenanthrenes. Except for a few examples of this eliminative photocyclization that were reported by other workers in 1974, this reaction has remained largely overlooked. Studies will be carried out with the parent o-methoxystilbene to establish optimum reaction conditions; in particular, the hypothesis will be tested that this eliminative photocyclization will proceed better in acidified methanol than it has in cyclohexane. Studies will be undertaken of the viability of this reaction as a method to overcome the regiochemical problems that sometimes limit the synthetic value of the well-known oxidative photocyclization method. For example, oxidative photocyclizations of meta-substituted stilbenes give mixtures of 2- and 4-substituted phenanthrenes, often with almost no regioselectivity. The eliminative photocyclizations of 2-methoxy-5-methylstilbene and 2-methoxy-3-methylstilbene will be carried out to test the hypothesis that these reactions will produce individual samples of 2- and 4-methylphenanthrene, respectively. Oxidative photocyclizations of 2-styrylnaphthalenes exhibit high regioselectivity, giving almost exclusively benzo[c]phenanthrenes; the regioisomeric benz[a]anthracenes are formed only in trace amounts. For similar reasons, the dibenz[a,j]anthracene and dibenz[a,h]anthracene systems also are not available by oxidative photocyclization. It is proposed to try to provide synthetic access to the benz[a]anthracene, dibenz[a,j]-anthracene, and dibenz[a,h]anthracene systems by carrying out eliminative photocyclizations of 3-methoxy-2-styrylnaphthalene, 2,4-dimethoxy-1,5-distyrylbenzene, and 2,5-dimethoxy-1,4-distyrylbenzene, respectively. These proposed experimental attempts to develop the synthetic utility of the eliminative photocyclization method focus on the simplest molecular examples, but the ultimate aim is to consider these examples as prototypes for future applications of the method to multi-substituted systems. The expansion of the capabilities for synthesizing polynuclear aromatic systems of the type considered here should have future importance for health-related studies because of the widespread occurrence of physiologically active natural products related to phenanthrenes, and the carcinogenicity of benzanthracenes.